Because the pool of ovarian follicles is set up during fetal life, the development of the ovary pre-determines female reproductive lifespan, including the time of onset of menopause. About 1-3% of all women undergo early menopause, either never going through menarche or stopping menstruation by the mid-30s, rather than reaching the standard reproductive lifespan of about 50. A fraction of such instances of early-onset premature ovarian failure (POF) is genetic. We have generated a mouse knock-out model for a gene that encodes a transcription factor, Foxl2, which that is mutated in some patients with early menopause (premature ovarian failure). We found that the earliest morphological anomalies involve a block in the formation of ovarian follicles, the fundamental units of ovary function. We then found that Foxl2-/- newborn follicle cells undergo female-to-male sex reversal, suggesting that dysregulation of sex determining genes even beyond the time at which they are usually thought to be active -- may be involved in ovarian failure in young and adult females. We moved on to provide evidence for a link between embryonic sex determining genes and premature ovarian failure. A notable example is seen in mice that we prepared lacking both Folx2 and Wnt4 (a double knock-out). In those mice, all cell types, including germ cells, reverse their sex during early development. This is the first experimental model of complete female-to-male sex reversal in mammals. We plan to continues to study the interaction of Foxl2 and male sex determining factors in the ovary, and have now identified arms of the pathways regulated by Foxl2 and Wnt4. Genes in those pathways will now be assessed for their role in vitro (organ cultures) and in vivo. In parallel, we are studying another transcription factor, Foxo3 that is also critically involved in the regulation of follicle dynamics and in premature ovarian failure in a mouse model. Toward this aim, we have generated transgenic mice over- or under-expressing Foxl2 and Foxo3 to test their effect on follicle fate and reproductive lifespan. In the past year we compared gene expression profiles associated with several time points during mouse ovary development using microarray and RT-PCR analysis. We found that some of the markers studied, in particular Foxl2 and Foxo3, seem to be required continuously to maintain the follicle cell sex fate and the follicle quiescence respectively. We performed a microarray analysis on gene expression profiles of our mouse models (single knock-outs Foxl2 -/- , Wnt4 -/- , Kit Wv/Wv;and double knock-outs Foxl2-/- Wnt4 -/-;and Foxl2 -/- Kit Wv/Wv) along with public datasets of several models of premature ovarian failure and ovarian dysgenesis (ablation of Emx2, Nobox, Lhx8, Foxo3 etc). Many genes are shared and active in both maturing gonads, consistent with the notion that the similarity predisposes the ovary to sex reversal toward testis. Foxl2-/- ovaries are closer to testis than their age-matched wild-type controls, and the time of convergence is even earlier in Kit-/-Foxl2-/- and Wnt4-/- compared to Kit-/- or wild-type ovaries, and is further anticipated in Wnt4-/-Foxl2-/- double knockout. We also compiled a list of candidate genes downstream of Foxl2. The list includes novel genes along with markers already known for their role in brain cortical patterning, vasculature function and gonadal steroidogenesis. We were also able to support the anti-testis role of Foxl2. A large fraction of the genes up-regulated in Foxl2-null ovaries was shared with testis (30-40%), further reflecting the partial sex reversal in absence of Foxl2. More specifically reflecting Foxl2 anti-testis activity, Sox9, Dmrt1 and Dhh, involved in testis formation, were up-regulated in its absence. In addition to testis genes, we identified some ovarian genes that were up-regulated in ovaries lacking Foxl2 or Wnt4. These represent novel candidates for an anti-testis action independent of Foxl2 and Wnt4. Overall, we have proposed a paradigm in which Foxl2 is required both for ovary differentiation and for the maintenance of female sex determination in the ovarian soma;and we also suggest that defects in Foxl2 and its pathway may be involved in a number of types of fertility problems. By cross comparing these data sets we were also able to identify a candidate gene that seems to be involved in the development of the cortico-medullary gradient of the ovary. An indispensable feature for the correct function of the ovary and for the fertility of the female, the gradient is lost in every case of ovarian dysgenesis. As for Foxo3 studies, it was already known that in its absence the entire cohort of follicles is derepressed for growth and the mouse ovary is left empty and sterile by 15 weeks. We now generated transgenic mice in which Foxo3 was specifically expressed only in the oocytes of primordial follicles. Furthermore, all phosphorylation sites were mutated so to maintain Foxo3 always nuclear and active. We found that at birth and even more at the time of follicle activation (7 dpn) markers involved in follicle formation and maintenance were up-regulated in transgenic mice and markers involved in follicle maturation were significantly down-regulated. Also the follicle count showed a striking difference in primordial follicle number and confirmed that the transgenic ovary is enriched in primordial follicles, which seem to remain in a quiescent state compared to wild-type. Microarray analyses of transgenic ovaries compared to Foxo3 knock-out and heterozygous ovaries of the same age confirmed that the presence of the transgene maintains the gonad in a significantly "younger" state. Extending the study to premenopausal, perimenopausal and menopausal ovaries confirmed the role of Foxo3 in maintaining the ovarian reserve and showed a possible role in prolonging the reproductive lifespan in females and retarding the aging processes of the ovary. In fact, transgenic aging ovaries had a larger number not only of primordial follicles, but also of all the other folliculogenesis intermediates, resulting in an ovary that was still active and possibly cycling.The mouse model thus showed dramatic and striking changes in follicle survival and maintenance.